Novel pharmacological applications of G-protein-coupled receptor–G protein fusions

Single, bi-functional polypeptides consisting of a G-protein-coupled receptor (GPCR) linked directly to a G protein α subunit have been employed for a number of years to study many aspects of signal initiation, including the roles of post-translational modifications, effects of mutations in both receptor and G protein and in the de-orphanisation of novel G-protein-coupled receptors. Recently, they have been used to improve signal-to-background in ligand assay screens and to study both agonist-directed signal trafficking and distinct conformational states of receptors. As well as such novel concepts in pharmacology, G-protein-coupled receptor–G protein fusions have recently been employed to examine receptor homo-dimerisation and hetero-dimerisation and are beginning to be used to explore allosteric effects within GPCR hetero-dimers.

Introduction

In humans, G-protein-coupled receptors (GPCRs) encoded by some 865 genes generate signals that are transduced into intracellular signals via the intermediacy of a family of hetero-trimeric guanine nucleotide binding proteins (G proteins), whose α subunits are encoded by less than 20 genes. Despite the relatively small number of G protein α subunit genes, at least 7 (Gα_s, Gα_i1, Gα_i2, Gα_i3, Gα_q, Gα₁₁, Gα₁₃) are commonly co-expressed by individual cells. This poses problems in the efforts to both observe and understand activation of a range of G proteins by a GPCR, particularly if this may occur selectively in response to different agonist ligands. Furthermore, individual G proteins are expressed to varying relative and absolute levels in different cells, and this creates challenges for pharmacologists who wish to understand aspects of GPCR–G protein interaction selectivity. In most cases, such challenges are approached by expressing a specific G protein α subunit in a cell system along with a GPCR, such that the G protein of interest becomes predominant in amount or has been modified (e.g. by the introduction of a pertussis toxin resistant mutation) [1, 2] such that simple steps can be taken to ensure that potential interactions of the GPCR with related endogenously expressed G proteins can be prevented. However, using such approaches, efforts to explore variations in agonist potency and efficacy for activation of two or more G proteins by a single GPCR remain problematic. This, partly, reflects that it remains nearly impossible to ensure that the expression levels and cellular distribution of the individual G proteins are equivalent and also that GPCR to G protein stoichiometry is, thus, constant for different pairings. Variations in these parameters must be anticipated to affect both apparent potency and relative efficacy of agonists [3].

Although first described more than 10 years ago [4, 5] and initially considered as little more than interesting curios, GPCR–G protein fusions (Figure 1) can overcome a number of these issues. Such fusions generally link the N-terminus of a G protein α subunit directly in-frame with the C-terminal tail of a GPCR to generate a bi-functional polypeptide within a single open-reading frame. As noted some time ago [6], the 1:1 stoichiometry between GPCR and G protein produced by their fusion should ensure a lack of receptor reserve and hence more direct measures of agonist affinity.

Over time, a wide range of GPCRs have been linked to G protein α subunits in such fusions, and except for G protein α subunits such as transducin 1, transducin 2 and gustducin, which have very restricted tissue distributions, virtually every mammalian G protein α subunit has been fused to one or more GPCRs [7•, 8••]. This has included the distribution restricted but highly GPCR-interaction promiscuous, G proteins Gα₁₅ and Gα₁₆, as well as artificial, chimeric G proteins [9] that are widely employed in ligand screening campaigns to direct agonist-mediated G-protein activation to specific intracellular read-outs. Although, initially, a potential concern, it does not appear that fusion results in inappropriate functional interactions between GPCRs and G proteins. For example, although effective in activating both Gα_s and a Gα_i1-Gα_s chimera, agonists at the IP prostanoid receptor were unable to activate Gα_i1 either when co-expressed as a separate polypeptide or when it was fused to this GPCR [10].

Earlier reviews [7•, 8••] have discussed many of the key studies in which GPCR–G protein α subunit fusions have been employed to investigate the role of information transfer of post-translational modifications, such as the addition of palmitic acid, to either a GPCR or a G protein. However, recent studies using this approach have established that for a α_2A-adrenoceptor–Gα_o1 fusion, though both GPCR and G protein become palmitoylated post-translationally, only for the G protein element the palmitoylation state is altered dynamically by agonist occupation of the receptor [11], whereas for an α_1B-adrenoceptor–Gα₁₁ fusion, agonist activation of the G protein is restricted when the G protein is de-palmitoylated but is unaffected by the palmitoylation status of the receptor [12].